ChatGPT's false memories, and would an inner critic stop AI hallucinations?

In episode 18 of The AI Fix our hosts discover that OpenAI's Advanced Voice mode is too emotional for Europeans, a listener writes a Viking saga about LinkedIn, ChatGPT is a terrible doctor, and the voice of Meta AI takes to Meta's platforms to complain about Meta AI reading things people post on Meta's platforms.

Mark discovers what Darth Vader really said on Cloud City, Graham rummages through ChatGPT's false memories, and our hosts find out why AIs need an inner critic.

Episode links:

• ChatGPT is awful at diagnosing medical conditions.
• OpenAI’s advanced voice mode is finally out…or is it?
• James McAvoy and Tom Brady fall for 'Goodbye Meta AI' hoax.
• Kristen Bell told Instagram to ‘get rid of AI’ before she became its official voice.
• Google has released two new versions of Gemini 1.5.
• "They stole my voice with AI."
• "The dark side of AI voice cloning (Elecrow responds)"
• OpenAI in throes of executive exodus as three walk at once.
• False memory.
• Examples and explanation of the Mandela Effect.
• The Mandela Effect: How False Memories Trick Your Brain Into Believing.
• Lost in the mall technique.
• Memory and new controls for ChatGPT.
• Hacker plants false memories in ChatGPT to steal user data in perpetuity.
• Spyware Injection Into ChatGPT's Long-Term Memory.
• Microsoft claims its new tool can correct AI hallucinations, but experts advise caution.
• The End of AI Hallucinations: A Big Breakthrough in Accuracy for AI Application Developers.

The AI Fix

The AI Fix podcast is presented by Graham Cluley and Mark Stockley.

Learn more about the podcast at theaifix.show, and follow us on Twitter at @TheAIFix.

Never miss another episode by following us in your favourite podcast app. It's free!

Like to give us some feedback or sponsor the podcast? Get in touch.

Our Sponsors:
* Check out Anthropic: https://claude.ai/aifix


Advertising Inquiries: https://redcircle.com/brands

Privacy & Opt-Out: https://redcircle.com/privacy